C li nic al E val u at i o n o f E a rl y
Cognitive Symptoms
J. Riley McCarten,
MD
a,b
KEYWORDS
Cognitive Dementia Memory Executive Alzheimer
KEY POINTS
An informant (spouse, family member) who knows the patient well is essential to obtaining
the history.
The onset, course, and nature of symptoms are the most important determinants of the
etiology of cognitive impairment.
Standard mental status tests are useful in clinic, but none are diagnostic. The physician
must have a good understanding of basic cognitive functions.
Targeting high-yield aspects of the neurologic examination allows an efficient evaluation
and contributes to the differential diagnosis.
The patient should be present throughout the process of taking the history and providing a
diagnosis.
INTRODUCTION
Evaluating symptoms of cognitive impairment shares the basic elements of evaluating
symptoms related to other complaints. The onset, course, and nature of symptoms
are key to the history and dictate the essential features of the examination. Although
it is important to allow patients/informants to describe symptoms in their own words, it
is essential that the physician clearly understands the symptoms from a medical
perspective. Furthermore, it is crucial that important questions relevant to the etiology
of symptoms are addressed. Just as for headache, chest pain, nausea, or dizziness,
the investigation of cognitive symptoms needs a combination of open-ended and
directed questions to elicit the critical features of the history.
Funding Sources: VA Cooperative Studies Program, VA HSR&D, NIH, Minnesota Veterans Medical Research & Education Foundation.
Conflicts of Interest: None.
a
Department of Neurology, University of Minnesota Medical School, 420 Delaware Street SE,
Minneapolis, MN 55455, USA; b Geriatric Research, Education and Clinical Center (GRECC),
Veterans Affairs Health Care System, One Veterans Drive, Minneapolis, MN 55417, USA
E-mail address: mccar034@umn.edu
Clin Geriatr Med 29 (2013) 791–807
http://dx.doi.org/10.1016/j.cger.2013.07.005
0749-0690/13/$ – see front matter Published by Elsevier Inc.
geriatric.theclinics.com
792
McCarten
Evaluating cognitive symptoms, however, does pose challenges not common to
other symptoms. Cognitive functions are by far the most complex of biological functions. Symptoms of neocortical dysfunction can be challenging for experts to describe,
much less the layperson. Patients often do not recognize the nature or severity of
cognitive symptoms, and may be defensive. Family and/or providers may minimize
the patient’s symptoms for a variety of reasons: Challenging the integrity of one’s
cognitive function may seem disrespectful, particularly if the investigation is uninvited.
Cognitive symptoms increase with age and may be seen as an inevitable part of aging,
are difficult to assess, and the evaluation may be unsatisfying. Lastly and most importantly, even moderately demented patients, several years into a progressive dementia,
may appear cognitively normal during a typical office visit. This fact cannot be overstated. Primary care physicians overlook cognitive impairment far too often.1,2
Though challenging, a structured approach to the evaluation of cognitive symptoms
should provide the information needed to make informed clinical decisions in the best
interest of patients, caregivers, and providers.
The foregoing describes the clinical evaluation in a medically stable outpatient. The
diagnosis of a progressive, irreversible cognitive disorder should not be made in an
acutely ill or stressed older adult. Conversely, awareness of underlying cognitive
impairment can prepare providers, family, and the patient for the commonly seen
exacerbation of cognitive symptoms when illness and stress do occur.
THE INTERVIEW
Overview
When the patient presents with cognitive complaints, or when problems are suspected during the course of the interview or examination, it is best to have a knowledgable informant available. Often this informant is in the waiting room. Because cognitive
problems are so common in the elderly, it is reasonable to ask any unaccompanied
older patient if he or she came with anyone, and if it would be permissible for that person to join the interview. The request is rarely denied.
If the patient is truly alone and cognitive symptoms are suspected, do not waste
time with a potentially unreliable history. Rather, move directly to mental-status
testing. If deficits are identified, the time saved can be used to identify an informant
and review the history. Problems identified in the course of a visit not specifically for
cognitive symptoms usually will require a separate appointment.
Symptoms of cognitive impairment often are intertwined with complex psychosocial
issues. When dementia is clearly present, cognitive impairment is readily recognized
as a contributing factor. When cognitive impairment is mild or questionable, it may
be difficult to identify the difference between symptoms caused by brain disease
and those related to anxiety, depression, medications, or situational factors. Patients
and families often have their own ideas about cause, and may offer elaborate but
tangential examples to support their viewpoints. Conversely, patients or, more often,
families may be reluctant to discuss sensitive issues. It is important, therefore, that the
physician establish the ground rules for the interview process to promote an efficient
and thorough assessment.
Setting the Ground Rules
When taking the history from more than 1 person (eg, a patient, spouse and/or adult
children), the patient and family members should be seated next to each other, not
across from one another. The physician should be able to see the reactions of family
members to the patient’s answers, and responses from family members often are
Early Cognitive Symptoms
guarded when the patient is looking at them. Family members also may try to coax or
coach the patient’s responses. Not only is it easier to maintain control over the interview if the physician can see the patient and all family members at the same time, more
information, often unspoken, is also gleaned.
Interviewing the Patient
The interview may be introduced by saying to the patient, “First I’d like to ask you
some questions and then, if it is OK with you, I’d like to ask your [family] some questions.” This request is rarely denied. It gives the patient the first chance to express concerns and promotes establishment of the doctor-patient relationship. Begin with an
open-ended question to the patient, such as “Why are you here today?” If the appointment is specifically to address cognitive symptoms, the answer may reveal important
information about the patient’s insight. If the patient does not volunteer cognitive complaints, ask specifically “How is your memory?” “Memory” in lay terms typically covers
a broad range of cognitive deficits. If deficits are acknowledged, ask the patient how
long he or she has been aware of it, and if it causes problems. Ask if the patient feels
clear-headed or confused, which may suggest delirium. Ask about mood and general
health. Usually this entire interview with the patient is brief.
Do not let family interrupt when interviewing the patient, or let the patient defer to the
family, noting, “They’ll get their chance.” Such interruptions or deferrals are telling, but
it is also important for the physician and family to hear what the patient believes the
problem to be. It may be useful to reiterate for the patient what you have heard.
Investigating the Family
With the patient’s permission, which is almost invariably granted, direct further questions to the family/informant. An important first inquiry is to establish the frequency of
contact the family member has with the patient. Sometimes the family member sees
the patient infrequently or has reestablished ties only recently. If someone who knows
the patient well did not attend the visit, find out why. Such information may not be volunteered but obviously is important.
Onset and Course
Ask of the informant, “Does [the patient] have a [cognitive] problem?” If yes, the onset
and course are critical to establishing a cause. Often, family members wish to emphasize recent changes that have prompted the current evaluation. Slowing evolving
changes that have caused problems in the recent past are very different from acute
or subacute changes that truly began only recently. A useful question is, “When
was the last time [the patient’s] thinking and memory was 100%?” Disregard reported
age-related changes. Patients and families, and often clinicians, may want to attribute
changes to “normal aging,” but such changes are rarely evident outside of a formal
neuropsychological assessment. Older adults are expected to manage complex medications, finances, and social relationships. Many are fully engaged in demanding careers. Changes resulting from disease are much more often attributed incorrectly to
“normal aging” than are changes of normal aging attributed incorrectly to disease.
The Major Causes of Brain Disease
The major causes of brain disease tend to have characteristic onsets and courses:
vascular events (stroke) and head trauma are acute; infectious/inflammatory conditions are subacute, usually evolving over hours to days and, rarely, weeks; neoplasms
are subacute, but typically evolve over weeks to months; toxic/metabolic conditions
are usually subacute, also over weeks to months, and fluctuate with the underlying
793
794
McCarten
cause; and degenerative brain disease has an indolent but inexorably progressive
course, with symptoms typically apparent for a year or more before an evaluation.
Because families often report changes as sudden, a useful question is, “Have you
ever thought he/she was having a stroke?”
Neurodegenerative Diseases
Neurodegenerative diseases, primarily Alzheimer disease (AD),3 but also diffuse Lewy
body disease (LBD)4 and frontotemporal dementias (FTDs),5 are the major causes of
cognitive disorders in older adults (Box 1). If symptoms are progressive but only gradually so, the longer ago they began, the more likely the primary underlying etiology is
neurodegenerative. When the onset is ambiguous but claimed to be recent, look for
evidence that symptoms were present further back in time. Clues include mistakes
in the patient’s management of his or her job, finances, medications, or other instrumental activities of daily living (IADLs; ie, the ADLs beyond basic self-care that are
needed to function independently). The further back in time the symptoms began,
the more likely the origin is neurodegenerative disease.
Nature of Symptoms
The nature of symptoms provides other important clues to the etiology of cognitive
impairment, and directed questions may be most useful for this purpose (Box 2).
Box 1
Common dementias and their frequencya
1. Alzheimer disease (60%–80%)
2. Diffuse Lewy body disease (5%–10%)
a. Dementia with Lewy bodies
b. Parkinson disease dementia
c. Multiple system atrophy
3. Frontotemporal dementia (FTD) (12%–25%)b
a. Behavioral variant
b. Language variant
i. Primary progressive aphasia—expressive aphasia
ii. Semantic dementia—receptive aphasia
c. Motor variant
i. Progressive supranuclear palsy
ii. Corticobasal degeneration
iii. FTD with parkinsonism
iv. FTD with amyotrophic lateral sclerosis
4. Vascular dementia (10%–20%)c
5. Mixed etiology (10%–30%)
a
Overlapping abnormalities are common.
Most FTD is young onset, <65 years old.
c
Vascular dementia is not consistently defined. In the absence of significant strokes involving
gray matter, the contribution of cerebrovascular disease to cognitive impairment is speculative.
Data from Refs.3,19,20
b
Early Cognitive Symptoms
Box 2
The history
Set the ground rules
Patient and family members next to—not across from—each other
Begin with patient but keep it brief. Do not allow family to interrupt
With the patient’s permission, direct questions primarily to family
Patient should be present throughout
Define onset and course or cognitive symptoms
When was the patient last completely normal?
-
Disregard “normal aging”
-
Onset is not when “things got bad”
Have symptoms progressed and, if so, gradually or suddenly?
-
Have you ever thought the patient was having a stroke?
Define nature of cognitive symptoms
Memory:
-
Repeating?
-
Misplacing?
-
Relying more on notes/calendars?
-
Forgetting names of familiar persons?
Language:
-
Trouble finding words?
Visuospatial/executive function:
-
Lost driving or other driving issues?
-
Mistakes with medications or finances?
-
Difficulty with former skills?
Preparing a meal
Household repairs
Using tools/appliances
-
Safety concerns?
Address associated behavior changes
Depressed, anxious, agitated?
Personality change?
-
Impulsive, inappropriate
-
Loss of empathy
Visual hallucination?
Paranoia?
Sleep disruption?
Involuntary movements or gait disturbance?
795
796
McCarten
Families and mildly impaired patients typically recognize symptoms with little or no
explanation from the examiner. Does he or she misplace? Struggle with names of
familiar persons? Rely more on notes and calendars? [typically reflecting memory deficits] Get lost driving? [memory or visuospatial deficits] Have trouble finding words?
[language deficits] Of course, all of these symptoms must reflect a decline from an
earlier baseline.
Patients are rarely aware of repeating themselves (a typical symptom even early in
the course of AD), but families typically readily recognize the simple question, “Does
he/she repeat?” Frequent repetition of questions, often verbatim by the patient, can be
exasperating for families.
Inquiries directed at executive functioning can be challenging. Executive functions
refer to skills typically localized to the frontal lobes and involve organizing, planning,
execution, and judgment. The ability to sequence, shift sets, and multi-task are all
dependent on executive functions, and deficits may cause difficulty preparing a
meal, completing a household repair, or preparing financial documents. Perhaps fortunately, executive dysfunction is often misinterpreted as a memory problem, because
patients appear to have forgotten how to do things. Executive deficits are common in
dementia, including AD, and are the hallmark of FTDs. A simple question that may
encompass executive dysfunction is, “Do you have safety concerns [for the patient]?”
Remarkably, despite the ready endorsement by families of many or all of the cognitive deficits described, it is often a surprise to them that the patient may have trouble
managing medications or finances. Even more remarkable, patients who are acknowledged to be dependent on others for virtually all IADLs, and even those needing assistance with basic ADLs, are thought by family to be safe drivers. Often, no family
members have ridden with or observed the patient’s driving.
Although it is important to return to driving or to other issues that relate directly to the
patient’s independence, do not let these matters sidetrack the evaluation.
Investigating Behavioral Symptoms
Behavioral changes are universal in brain disorders that cause cognitive impairment.6
Asking the family about the patient’s mood is important, but one must be aware that
vegetative signs reflecting apathy, not depression, may accompany even early symptoms of dementia. It may take a skilled geropsychiatrist to differentiate the apathy of
dementia from true depression that often accompanies dementia.7 Older adults with
cognitive disorders may improve significantly with treatment of their comorbid
mood disorder, but depression, anxiety, mania, and other psychiatric conditions rarely
develop as a primary disorder in late life. Even in the absence of overt cognitive impairment, such patients should be watched for an evolving neurodegenerative disorder.
Personality Change
Many changes in behavior may be interpreted as a personality change, but true personality change, characterized by impulsive, odd, and inappropriate behavior, apathy and
indifference, or coarsening of affect with loss of empathy are hallmarks of behavioral
variant FTD and also can be seen in traumatic brain injury (TBI). Such patients may
appear quite normal in the clinic and may test well on structured mental-status examinations. Often, only the people who know them best recognize the change. When the
patient appears normal and the family appears distressed, think frontal lobes.
Hallucinations and Delusions
Visual hallucinations are relatively specific to LBD (synucleinopathies),8 particularly
early in the disease course, and should be asked about in all older adults with
Early Cognitive Symptoms
suspected cognitive impairment. Auditory hallucinations are much less specific and
are not common in the early course of dementia. Delusions, which are common in dementia, may be reported by the family as hallucinations based on the patient’s statements of having seen or heard things. Confirm that family members have observed the
patient hallucinating and are not just reporting what the patient has said.
Patients may be more prone to misinterpret sights or sounds (illusions), sometimes
attributing a threatening quality to them. Forgetfulness and impaired judgment also
can trigger paranoid delusions and may prompt accusations of stealing or other
bad behavior. Pleasant delusions, such as the patient believing he or she recently
visited with a long-dead friend, also may occur and, while sometimes upsetting to
families, are not harmful and may actually offer opportunities to engage the patient
in conversation.
Sleep
Sleep changes are common in older adults, and disrupted sleep can exacerbate
cognitive symptoms. However, it is uncommon for sleep disorders to be the primary
cause of cognitive impairment. The treatment of sleep apnea, restless legs, or other
disorders causing excessive daytime sleepiness can improve attention but rarely eliminates the perceived changes in cognitive impairment. Rapid eye movement behavior
disturbance, a sleep disorder characterized by often dramatic motor activity accompanying vivid dreams, is suggestive of LBD,9 as are visual hallucinations.
Motor Disturbances
Involuntary movements and/or a gait disturbance may suggest LBD, a motor variant
FTD, or, in another context, a toxic encephalopathy (delirium). Lateralized weakness
or numbness may suggest focal brain lesions, such as stroke, subdural hematoma,
or TBI. Bowel or bladder disturbance or symptoms of presyncope may indicate dysautonomia, also seen in LBD (multiple system atrophy).
The Past Medical History
A complete medical history is important, with particular emphasis on prior neurologic
or psychiatric problems. In general, the brain is a resilient organ which, given enough
time, can compensate for significant impairment to other organs. Apart from endstage pulmonary, cardiac, hepatic, or renal disease, patients have at least intermittent
mental clarity.
Medications
A review of medications with special attention to adherence and to drugs with central
nervous system activity, including over-the-counter and illicit drugs, is important. Stable doses over years of benzodiazepines, opiates, antiepileptic drugs, or even alcohol
are not likely to present as a progressive cognitive impairment unless blood levels are
increasing and/or tolerance is decreasing. Not infrequently, patients with cognitive
impairment make medication errors. In addition, common behavioral and psychological symptoms of dementia may lead to self-medication with drugs or alcohol.
The Family History
The family history is important, particularly if there is a strong history of mid-life neurodegenerative disease. The vast majority of dementia is sporadic, and late-life dementia in a relative condones only a mild increase in risk. Nevertheless, it is often a concern
that prompts the patient to seek an evaluation.
797
798
McCarten
The Social History
The patient’s education, occupation, current activities and hobbies, home life, social
support, and belief systems are all important factors that influence the presentation of
cognitive symptoms.
The Review of Systems
Though often mandatory, the review of systems from an unreliable historian is virtually
worthless. A standard form that the patient can review with a knowledgable family
member before the visit to the clinic is preferred.
Interviewing the Family Separately
Although families often want to speak to the physician independently, it is best to
address issues of cognitive impairment as openly and honestly as possible with the
patient present. Beyond the ethical issues, deceiving or withholding information
from the patient may sabotage the doctor-patient relationship. Patients may be forgetful, but they usually recognize whom they trust. Moreover, when families are reluctant
to speak in front of the patient, the patient typically has limited insight and fairly significant cognitive impairment. In such situations, families are usually unduly concerned
about what the patient will remember. The patient often is indifferent or, if upset,
has a short-lived irritation. For patients who are more mildly or even questionably
impaired, there is rarely justification for discussing concerns with the family without
the patient being present.
The process of diagnosing dementia may be stressful for patients, families, and
physicians, but trying to spare patients from potentially bad news is a poor strategy
for managing any disease. Usually, when the physician is straightforward families
do not feel the need to speak apart from the patient, and both patient and family
are grateful for the physician’s honesty.
THE EXAMINATION
Background
Although a general medical examination is important, the neurologic examination is
key to diagnosing cognitive impairment. Observations about the patient’s psychological state, including mood, affect, character of speech, thought content, and insight
may influence the assessment, but in the cooperative and attentive older adult, cognitive impairment usually can be detected unless psychiatric symptoms are extreme.
The Neurologic Examination
The neurologic examination uses a carefully constructed approach to localize lesions
within the nervous system. Findings are addressed in a deliberate fashion, beginning
at one end of the neural axis (brain cortex) and sequentially adding considerations of
other nervous system components. The mental status, reflecting function of the cerebral hemispheres, is considered first. The cranial nerves add a consideration of the
brainstem. The motor examination adds the spinal cord, efferent nerves, neuromuscular junction, and muscle. The sensory examination adds afferent nerves and
ascending spinal cord pathways that eventually relay in the thalamus and project to
cortex. Lastly, coordination, station, and gait testing adds a consideration of cerebellar and basal ganglia input, which work through the motor system and depend
on sensory feedback.
Early Cognitive Symptoms
Localizing Cognitive Functions
In the alert (eyes open) and attentive (maintains eye contact) adult, the major spheres
of cognition are memory, language, and visuospatial and executive function (Fig. 1).
Each localizes to important brain regions.
The critical structures for making new memories are the hippocampi, deep in the
medial temporal lobes. Language localizes to the dominant (usually left) temporoparietal cortex, visuospatial function to a similar region of the nondominant hemisphere,
and executive function to the frontal lobes (Fig. 2). Note that language, visuospatial
function, and executive function are neocortically based, whereas memory depends
on the much more primitive archicortex of the hippocampus. Neorcortical deficits
may be difficult to identify in the clinic, particularly in a well-educated patient. Recent
or “short-term” memory, the crucial ability to learn and remember new information,
tends to be easier to test and is usually affected first and foremost in AD.
Orientation is not useful for localization, is not sensitive to cognitive dysfunction, and
has no uniformly recognized criteria. If a patient presents at the correct time and place,
little is gained by quizzing orientation.
The Mental-Status Examination
The initial mental-status examination should be administered in a rigorous fashion,
ideally by a nurse or other support staff trained in the administration of standard tests.
Many such tests are available.10,11 If there is no reason to expect cognitive dysfunction, a brief (w2 minutes) screen, such as the Mini-CogÔ,12 is adequate. If problems
are suspected, a more extensive test, such as the Montreal Cognitive Assessment
Fig. 1. Structures that are midline and deeper in the brain are more primitive. After recovery
from an acute, bilateral insult to the brain (eg, anoxia), alertness requires only the reticular
activating system. Attentiveness requires thalamic drivers to keep the cortex in a receptive
state. Recent or “short-term” memory is dependent on the hippocampi, a primitive archicortex. Language, visuospatial, and executive functions are all more evolved, depend on the
neocortex, and may be difficult to assess in a brief examination.
799
800
McCarten
Fig. 2. Neocortical structures, and particularly the large frontal lobes, are the most evolved
and dynamic part of the brain. Only about 10% of neocortex is primary sensory (vision, hearing, touch, smell, taste) or motor (output) cortex. The vast areas of association cortex create
the individual’s perception of the world, act on those perceptions, and otherwise “think.”
Though critical to the individual’s function, they may be challenging to test in the clinic.
(MoCA),13 can be used. The scores on such tests are not diagnostic but do alert the
physician to the potential need for further evaluation.
It is essential for the physician have a good working knowledge of cognitive function
to help decide if (1) the patient is a reliable historian and responsible partner in recommended cares; and (2) further testing is needed and, if so, what.
Families should be invited to observe the mental-status examination. Not infrequently, they are shocked at the patient’s performance.
Recent Memory
Because of its importance to the history and treatment recommendations, it is imperative that the recent or short-term memory be assessed. A patient with impaired
recent memory is, de facto, an unreliable historian and also should not be accountable
for adhering to treatment recommendations, including medications. Just as with any
other abnormal examination finding, the physician should reassess recent memory
as needed until he or she is convinced that there is or is not a problem.
Recent memory is tested by delayed recall. Most commonly the patient is asked to
remember 3 unrelated words and to later recall the words after an interference (distracter) task. Not all words lists are equal,14 and a standard list, such as “leader, season, table” should be used. The interference task, not the time delay, is key to
identifying deficits in recent memory and must be adequately demanding to fully
engage the patient. Common interference tasks include serial 7 subtractions (“Count
backwards from 100 by 7s”) and reciting the months of the year in reverse order, either
of which is also a good test of attention/concentration. The clock draw task, the interference task used in the Mini-CogÔ, assesses visuospatial and executive function
(see later discussion).
If a patient struggles to recall the 3 words after the first interference task, have the
patient repeat them again, do another interference task, and again test recall. Struggling a second time is concerning. Struggling a third time, again rehearsing the words
and using a different interference task, is very likely not normal. The test can be refined
by providing letter or category cues, then multiple choice if needed. Patients who do
Early Cognitive Symptoms
not recall the words but readily get them with a simple cue (eg, “Starts with the letter,”
“A piece of furniture”) may be intact, particularly if they later recall the words without
cues. Patients who cannot recognize the words from a multiple-choice list (eg, “sugar,
sailor, or season,” or “chair, table, or sofa”) are most concerning, particularly on
repeated testing.
Well-educated patients may be able to bear down enough to appear intact on a simple quiz, but deficits may be revealed on their knowledge of current events. Major
recent events, such as elections and natural catastrophes, or historic events, such
as 9/11, should be recalled by anyone who acknowledges following the news. Mistakes, particularly when unrecognized by the patient, may be startling.
Language
Even when word-finding difficulties are reported, they may not be apparent when talking with the patient. Asking the patient to follow multiple step instructions—essentially,
3-stage commands—is a simple test of comprehension and can be incorporated into
most examinations. Be deliberate in giving the patient multiple instructions at once,
without cues, as opposed to a series of 1-step instructions. For example, say,
“Hold your hands straight out in front of you, elbows straight, palms up, at shoulder
level, and close your eyes” [instructions for testing drift]. Decide what is important
for any routine examination and invent a way to incorporate a multistage command
into that examination. This approach represents an efficient use of time and reveals
something of the patient’s likelihood of being able to follow advice.
Generative Naming
Generative naming may be revealing, even when conversation and comprehension
appear intact. Semantic (category) fluency tends to reflect temporal lobe functions,
whereas phonemic (letter) fluency taps into frontal lobe function.15 Patients are given
1 minute to generate as many names, typically animals, or words beginning with a specific letter, typically f, a, or s. Jot down the starting time to the second, as tracking
words can be demanding. Normally patients generate at least 11 words per minute.
Patients with AD may have more difficulty with categories, whereas FTD patients
may have trouble with letters.
Naming objects is difficult to qualify and quantify in the clinic, and may be intact until
deficits are relatively severe. It is also less revealing than difficulties following instructions or generating words.
Be aware that intact language skills do not preclude significant deficits in memory,
visuospatial function, or executive function. Often, even severely cognitively impaired
patients appear normal because of preserved language and social skills, basically
“talking a good game,” and fooling both family and providers.
Visuospatial Skills
Visuospatial skills are difficult to test in the clinic, despite the profound impact such
deficits may have on function. Their assessment often overlaps with that of executive
function. Copying intersecting pentagons, in which 2 5-sided figures form a 4-sided
intersection, or copying a cube such that the figure appears 3-dimensional, the appropriate lines are parallel and of equal length, and no lines are added, are relatively specific bedside tests for visuospatial deficits.
Executive Function
Executive functions are the most evolved and complex of biological functions in the
known universe. Almost half of the brain—the frontal lobes—is dedicated to executive
801
802
McCarten
functions. These functions also may be most important to successful independent
living, because they are the basis of planning, organizing, recognizing patterns, shifting attention as needed, and making sound judgments. Nevertheless, deficits may be
difficult to recognize in the clinic. Because they are so involved in interpersonal interactions, it is not surprising that persons who know the patient best may recognize
changes before others. Facial expression, body language, and tone of voice may
speak volumes to family members but mean little to a relative stranger.
It must be reiterated here that when a patient appears normal and has no complaints, but the family looks distressed, think frontal lobes.
The Clock-Draw Task
The clock-draw task taps into both visuospatial and executive functions (planning,
abstracting), and is an efficient way to screen for either. There are a variety of ways
to score the clock. Basically all 12 numbers should be present, in order, only once,
in the correct clockwise direction, and the hands should point to the correct numbers.
Ideally the drawing should be well planned, with quadrants defined, numbers correctly
spaced, and hands of the appropriate length. The time “11:10” is most often used, the
instruction being, “Place the hands at 11:10, 10 minutes past 11.” Impaired patients
are often “stimulus bound,” and tend to draw hands pointing to the 10 and 11. Other
times, including 8:20 and 1:45, also require abstraction and are only slightly less likely
to reveal deficits.14
Sequencing
Sequencing may be challenging for patients with executive dysfunctions, if not more
global cognitive impairment.16 The Luria sequencing task, or fist-edge-palm test, is
administered as follows. Instruct the patient to “Watch me.” Do not provide any verbal
cues. Demonstrate the sequence of fist (fingers balled, perpendicular to the floor),
edge (hand open, perpendicular to the floor), palm (hand open, parallel to floor), tapping your hand on your thigh as you perform each movement. Use a silent 4-count
beat (1 beat rest after each sequence) to clearly distinguish the 3 steps. Perform the
demonstration 3 times, and then say to the patient, “Now you do it.” A normal patient
performs this readily, 3 times in a row. If the patient does not perform it correctly 3
times, say “Do it with me,” and demonstrate 3 more trials, making sure the patient
mimics your hand position. Then say “Keep going.” The test is scored as spontaneously correct, correct with coaching, partially correct (1 or 2, but not 3 correct sequences in a row), or unable to do.
Set-Shifting
Set-shifting is another challenging task for persons with frontal lobe dysfunction, and
also for those with any significant deficits in attention (delirium). Trails B is a neuropsychological test whereby numbers and letters are randomly distributed on a page and
the patient must connect them in the correct, alternating sequence: 1-A-2-B-3-C, and
so forth.
Trails B can be administered orally, tapping into executive function, attention, and
working memory. Say to the patient, “Complete this sequence: 1-A-2-B-3-C.” Most
can complete the entire sequence, arriving at 26-Z or close to it. Those with executive
dysfunction often charge ahead but quickly lose track of the sequence, responding
incorrectly with all numbers or letters, or sequences that are clearly out of order. Those
with even subtle delirium, (eg, mild intoxication or medication side effects) may quickly
lose their place. Patients able to complete this task are unlikely to have delirium, and
any other cognitive deficits identified are usually the product of brain disease.
Early Cognitive Symptoms
The oral Trails B task is also an excellent interference task for testing memory. It is
virtually impossible to rehearse newly presented information, thereby enhancing
delayed recall, and correctly attend to this task.
As noted under language function, phonemic fluency (the number of words generated in 1 minute beginning with f, a, or s) also tends to be disproportionately impaired
in disorders affecting the frontal lobes.
Other clinical assessments of executive/frontal lobe function often have biases and
are difficult to quantify, but can be revealing to the experienced clinician. The interpretation of proverbs depends on the patient’s age, culture, and education. The response
demonstrating the optimal abstraction is another proverb of the same meaning. Interpreting similarities and differences (eg, how are an orange and a banana alike?) may
be complicated by creative responses (both are good sources of potassium).
The Remainder of the Neurologic Examination
The neurologic examination is often seen as overly complex and time consuming, but
most findings relevant to the differential diagnosis of cognitive impairment can be
gathered in short order (Box 3).
It is much better to make a few valid, reproducible observations than to make generalizations such as “grossly intact,” “nonfocal,” or “nonlateralizing.” Each of the
following descriptions of a normal examination take only moments to observe or
test, yet are of high yield in terms of assessing the integrity of the nervous system
outside of the mental status.
Cranial nerves: Visual fields full, extraocular movements (EOMs) intact, face symmetric, speech articulate, hearing intact to soft voice.
Stands easily from a chair without pushing off, gait normal, walks well on tiptoes and
heels, station normal, postural reflexes intact, Romberg negative, Drift negative,
finger-to-nose testing and rapid alternating movements intact.
In the most abbreviated neurologic examination, apart from always testing recent
memory, the EOMs and the gait are the most important to test. Develop skill in watching for smooth, seamless pursuit movements without evidence of intrusion saccades.
The observation of a normal gait, which is readily recognized, may be an invaluable
piece of information to document and track.
ADDITIONAL TESTING
Brain Imaging
Frequently, brain imaging (magnetic resonance imaging or computed tomography) is
used as a substitute for the neurologic examination. The brain is tremendously dynamic,
and a static picture may reveal little about neurologic function. The most common
cognitive disorders, namely neurodegenerative dementias, are not evident on routine
imaging, and functional imaging is confirmatory at best, not diagnostic. The strongest
argument for brain imaging for cognitive impairment is that the patient is typically an unreliable historian, and a significant injury or event may be forgotten. Many “abnormal”
findings on imaging are nonspecific (eg, the ubiquitous “small-vessel ischemic disease,” which may or may not represent vascular disease) and of no clinical significance.
Imaging does rule out significant cortical strokes, TBIs, tumors, subdural hematomas,
and hydrocephalus, any of which could cause or aggravate cognitive symptoms.
Laboratory
Vitamin B12 and thyroid-stimulating hormone are the only laboratory tests that
commonly are not already part of the clinical laboratory data. A complete blood count,
803
804
McCarten
Box 3
High-yield neurologic examination findings in the evaluation of cognitive symptoms
Cranial Nerves
Extraocular movements (cranial nerves III, IV, VI)
Rationale: Assesses the most discrete of motor movements, requiring multiple coordinated
areas of brain function. Abnormal in:
-
Progressive supranuclear palsy, Parkinson disease
-
Some structural lesions
-
Drug toxicity
-
All brain disease, at least subtly
Procedure: “Follow my finger with your eyes”
-
Move finger to extremes of gaze at varying speeds
-
Observe particularly for smoothness of pursuit movements
-
Do not hold finger too close, as difficulty converging (older adults) may produce
nonpathologic intrusion saccades (choppy eye movements)
Normal: Seamless, conjugate pursuit of target
Abnormal:
-
Dysconjugate or lack of full range of motion ( mildly limited up gaze in older adults)
-
Significant nystagmus (beyond mild, symmetric end point)
-
Decrease in smoothness of pursuit movements (nonspecific but sensitive)
Speech (cranial nerves V, VII, IX, X, XII)
Rationale: Multiple brain and peripheral structures are involved. May be abnormal in:
-
Bilateral corticobulbar (pseudobulbar) lesions
-
Bulbar (lower brainstem) lesions
-
Basal ganglia and cerebellar disorders
-
Some lateralized structural lesions
-
Drug toxicity
Procedure: Assess speech quality during interview
Normal: Speech articulate, volume normal
Abnormal:
-
Dysarthria (slurred; eg, cerebellar/brainstem lesions, drug toxicity)
-
Hypophonia (eg, Parkinson disease)
-
Halting (eg, basal ganglia, language cortex)
-
Dysphonia (hoarseness)/nasal speech (brainstem or peripheral structures)
Drift
Rationale: May detect lateralized motor, sensory, cerebellar deficits; highly reproducible
Procedure: Arms extended, palms up, shoulder level, eyes closed
Normal: Holds for 10 seconds; tremor is discounted
Abnormal: One arm
Pronates, may also drift downward (corticospinal tract lesion)
Drifts laterally to 45 (ipsilateral cerebellar dysfunction)
Drifts upward (may be central sensory deficit)
Early Cognitive Symptoms
Stand
Rationale: Detects weakness, postural instability that poses risk of falls; highly reproducible
Procedure: Cross arms over chest and stand up
Normal: Patient stands easily without pushing off
Abnormal:
Struggles but stands without pushing off
Must push off to stand
Needs assistance to stand
Cannot stand
Gait
Rationale: Requires integrated motor, sensory, basal ganglia, cerebellar function. Gait is
often abnormal in:
Lewy body disease
-
Untreated Parkinson disease
-
Dementia with Lewy bodies
-
Multiple system atrophy
Tau disorders
-
Progressive supranuclear palsy
-
Corticobasal degeneration
-
Other frontotemporal dementias
Huntington disease
Post–cardiovascular accident or traumatic brain injury
Normal pressure hydrocephalus
Cerebellar or toxic/metabolic disorders
Procedure: Tell patient, “Take a fast walk down the hall [to bring out abnormalities].” Then,
“Come on back [to observe turns].” Note:
Base
Stride
Posture
Arm swing
Turns
Normal: Walks normally 30 ft down the hallway and returns
Abnormal:
Hemiplegic/paretic (lateralized brain lesion)
Parkinsonian (basal ganglia, relative dopamine deficiency)
Choreiform (basal ganglia, relative dopamine excess)
Ataxic (cerebellar; alcohol intoxication causes classic gait ataxia)
Magnetic/apractic (normal pressure hydrocephalus and frontal lobe disorders)
Neuropathic (peripheral nerve; steppage gait in extreme cases)
Myopathic (muscle; may cause waddling with weakness of pelvic girdle)
Antalgic (nonneurologic gait disturbance related to pain)
805
806
McCarten
liver and renal function tests, glucose, electrolytes, and calcium should be documented. When appropriate, testing should be done for sexually transmitted disease
(rapid plasma reagin, human immunodeficiency virus). The onset, course, and nature
of symptoms may dictate further evaluations (eg, Lyme titers, drug and heavy metal
screens) but are infrequently indicated and should not be pursued in what appears
to be a typical case of AD, LBD, or FTD.
Neuropsychological Testing
Formal neuropsychological testing is invaluable when assessing truly mild or questionable symptoms of cognitive impairment. Each of the major domains of cognition
already outlined is assessed with multiple, standardized tests. The neuropsychologist
with an interest in older adults may be skilled at diagnosing AD versus LBD versus
FTD. Not all neuropsychologists have the same skill sets, and some are much less
willing than others to endorse a pattern suggesting an irreversible, progressive brain
disease. The neuropsychologist may place undue weight on the report of neuroimaging (small-vessel ischemic disease), vascular risk factors, and a history of alcohol use.
Functional Assessments
Occupational therapists may be skilled at assessing ADLs and IADLs in older adults. A
structured, performance-based evaluation, such as the Cognitive Performance Test,17
has the advantage of allowing family members to observe as patients are directed to
perform various tasks. Prompts are provided in a graded fashion until the task is
completed, revealing the degree of assistance needed. Such information may be
crucial in deciding if the patient has the “impairment in function” that is integral to
the definition of dementia, and can be telling about the safety of the patient’s current
activities and environment.
Other Tests
When indicated, a sleep study should be pursued before neuropsychological testing.
An electroencephalogram occasionally is helpful if there is a true question of spells of
confusion, or if a subacute progression suggestive of prion disease is suspected.
It is important to recognize that the vast majority of cognitive disorders are clinical
diagnoses, and ancillary tests rarely identify a reversible cause.18
THE ASSESSMENT
The physician’s first job is to decide if there is: (1) no cognitive impairment; (2) cognitive
impairment, not dementia; or (3) dementia. For either of the latter 2 possibilities a specific etiology should be sought, which will directly relate to the prognosis and steer the
patient and family to the resources needed.
REFERENCES
1. Valcour VG, Masaki KH, Curb JD, et al. The detection of dementia in the primary
care setting. Arch Intern Med 2000;160:2964–8.
2. Chodosh J, Petitti DB, Elliott M, et al. Physician recognition of cognitive impairment: evaluating the need for improvement. J Am Geriatr Soc 2004;52:1051–9.
3. Plassman BL, Langa KM, Fisher GG, et al. Prevalence of dementia in the United
States: the aging, demographics, and memory study. Neuroepidemiology 2007;
29:125–32.
Early Cognitive Symptoms
4. Schneider JA, Arvanitakis Z, Yu L, et al. Cognitive impairment, decline and fluctuations in older community-dwelling subjects with Lewy bodies. Brain 2012;135:
3005–14.
5. Baborie A, Griffiths TD, Jaros E, et al. Pathological correlates of frontotemporal
lobar degeneration in the elderly. Acta Neuropathol 2011;121:365–71.
6. Casanova MF, Starkstein SE, Jellinger KA. Clinicopathological correlates of
behavioral and psychological symptoms of dementia. Acta Neuropathol 2011;
122:117–35.
7. Wright SL, Persad C. Distinguishing between depression and dementia in older
persons: neuropsychological and neuropathological correlates. J Geriatr Psychiatry Neurol 2007;20:189–98.
8. Burghaus L, Eggers C, Timmermann L, et al. Hallucinations in neurodegenerative
diseases. CNS Neurosci Ther 2012;18:149–59.
9. Zanigni S, Calandra-Buonaura G, Grimaldi D, et al. REM behaviour disorder and
neurodegenerative diseases. Sleep Med 2011;12(Suppl 2):S54–8.
10. Ashford JW. Screening for memory disorders, dementia and Alzheimer’s disease.
Aging Health 2008;4:399–432.
11. Brodaty H, Low LF, Gibson L, et al. What is the best dementia screening instrument for general practitioners to use? Am J Geriatr Psychiatry 2006;14:391–400.
12. Borson S, Scanlan JM, Chen P, et al. The Mini-Cog as a screen for dementia: validation in a population-based sample. J Am Geriatr Soc 2003;51:1451–4.
13. Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal cognitive assessment,
MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc
2005;53:695–9.
14. McCarten JR, Anderson P, Kuskowski MA, et al. Screening for cognitive impairment in an elderly veteran population: acceptability and results using different
versions of the Mini-Cog. J Am Geriatr Soc 2011;59:309–13.
15. Tupak SV, Badewien M, Dresler T, et al. Differential prefrontal and frontotemporal
oxygenation patterns during phonemic and semantic verbal fluency. Neuropsychologia 2012;50:1565–9.
16. Weiner MF, Hynan LS, Rossetti H, et al. Luria’s three-step test: what is it and what
does it tell us? Int Psychogeriatr 2011;23:1602–6.
17. Burns T, Mortimer JA, Merchak P. Cognitive performance test: a new approach to
functional assessment in Alzheimer’s disease. J Geriatr Psychiatry Neurol 1994;7:
46–54.
18. Clarfield AM. The decreasing prevalence of reversible dementias: an updated
meta-analysis. Arch Intern Med 2003;163:2219–29.
19. Holsinger T, Deveau J, Boustani M, et al. Does this patient have dementia? JAMA
2007;297:2391–404.
20. Daviglus ML, Bell CC, Berrettini W, et al. NIH state-of-the-science conference
statement: preventing Alzheimer’s disease and cognitive decline. NIH Consens
State Sci Statements 2010;27:1–30.
807